Colorectal cancer represents a major public health issue but also provides exceptional opportunities for investigating human tumorigenesis. We have recently developed experimental approaches that allow us to analyze human cancers in unprecedented detail and address questions that were difficult or impossible to answer in the past. We intend to use these approaches to fill gaps in our understanding of the early events in colorectal tumor development. Aim #1 is directed at identifying the genetic basis of two polyposis syndromes, Serrated Polyposis Syndrome (SPS) and Atypical Adenomatous OligoPolyposis (AAOP). To determine the genetic basis for both these syndromes, we will perform whole genome sequence analysis of normal and tumor tissues from representative patients. Aim #2 is directed at identifying the clonal relationship and timing of genetic alteratios during colorectal tumorigenesis in the canonical Adenoma-Carcinoma and the Serrated Colorectal Cancer Pathways. We will perform genome-wide analyses of clonal somatic mutations in a unique collection of lesions including hyperplastic polyps, serrated adenomas, sessile serrated adenomas, small adenomatous polyps and, most importantly, invasive and adjacent non-invasive components (adenomatous and hyperplastic) from the same lesions. We will thereby be able to determine the clonal relationships among different components of the same tumor and evaluate the time periods required for tumor progression. Aim #3 is directed at characterizing non-clonal somatic mutations in normal and neoplastic colorectal tissues. The number and type of non-clonal somatic mutations can provide a window into past endogenous and exogenous mutagen exposures. We have recently developed a technique (SafeSeqS) that can reliably identify non-clonal somatic mutations in clinical samples. SafeSeqS will be used to assess such mutations in benign and malignant tumors as well as in normal colorectal tissues. Aim #4 is directed at identifying the full compendium of early events in colorectal tumorigenesis. Several lines of evidence support the idea that APC or CTNNB1 mutations are the initiating events in colorectal tumorigenesis. Is tumor development simply gated by inactivation of the APC pathway or are additional early changes required? Using technologies developed during the past funding period, we will pursue an integrated (i.e., genetic, epigenetic and transcriptomic) analysis of multiple synchronous adenomas removed from the same patients to look for recurring genetic and epigenetic changes that may act as early events in tumor development. Relevance: Our specific aims are directed at identification of the early events in colorectal tumorigenesis. It can be argued that understanding these early changes is essential for the development of more effective strategies for colorectal tumor prevention and treatment. PUBLIC HEALTH RELEVANCE: Colorectal cancer is the second leading cause of cancer deaths in the United States with 141,000 cases and 49,000 deaths expected in 2011. Understanding the earliest changes in colorectal tumorigenesis will be key to developing the most effective preventions and treatments for colorectal cancer. The goals of the studies proposed in this application are to characterize early events in various types of colorectal tumors.